This research will examine the relationships between physical health (primarily mortality, secondary morbidity) and certain key social and emotional predictors. In particular, this research will test hypothesized relations between major social stressors/emotional reaction patterns earlier in life and the subsequent development of illness in middle and old age. An important secondary aim is the analysis of sex differences in longevity, as influenced by these same psychosocial variables. The design is an archival prospective cohort study, that is, a longitudinal study involving a rich existing data set. The raw data come from the Gifted Children Study, begun in 1921 by L. Terman and continued by R. Sears and others. The subjects are 857 men and 671 women followed from their childhood in the 1920's until the 1980's, the longest continuous cohort study ever conducted. The initial study of the research involve a culling of the relevant variables from more than 4000 raw variables collected on the subjects, the creation of psychometrically sound variables and indexes, and the examination of the interrelationships among the variables. The later stages of the research involve the prediction of health and longevity as a function of psychosocial stresses and resources --including objective losses, emotional temperament, personality, and social support. The primary "hard" endpoint is date of death (longevity). Key unanswered questions in psychosocial epidemiology, community medicine, and health psychology will be addressed using survival analyses, supplemented by logistic regression analyses as appropriate. As part of the analysis, sex differences in health and longevity at different stages of middle and old age will be examined in an attempt to shed light on puzzling sex differences that have previously been documented. This study will thus provide important information relevant to our understanding of the influence of individual and social factors on morbidity/mortality across the life span.